CA2155303C - Treatment for insulin dependent diabetes - Google Patents

Abstract

A method for the prevention of insulin dependent (type I) diabetes. The method comprises administration of an antibody, polypeptide or other molecule recognizing VLA4.

Description

TREATMENT FOR INSULIN DEPENDENT DIABETES
FIELD O! 'PM I1QVEliT2OI/
The present invention relates to a treatment for insulin dependent (type-I) diabetes. More particularly, this invention relates to the use of antibodies recognizing the integrin vLA4 (yery late ,Antigen 4) in the prevention of diabetes.
HACRaAOUND OF TEE INPEN'PIOL1 Insulin dependent diabetes (also termed type-I
diabetes and formerly juvenile onset diabetes mellitus) has been classified during the past two decades as a chronic autoimmune disease. In this disorder, cells producing insulin (p cells) within the pancreatic islets are selectively targeted and destroyed by a cellular infiltrate of the pancreas. This inflammatory infiltrate affecting the islets has been termed insulitis. Cells producing insulin comprise the majority of islet cells but less than 2% of the total pancreatic mass (Castano and Eisenbarth, 1990, [1];
Fujita et al., 1982 (2); Foulis et al., 1986 13]). The development of type I diabetes can conceptually be divided into six stages, beginning with genetic susceptibility and ending with complete p cell destruction (Eisenbarth, 1986 [4]). Stage I is genetic susceptibility, which is a necessary but insufficient condition for development of the disease. A
hypothetical triggering event (Stage II) leads to active autoimmunity against p cells (Stage III). In Stage III, the p cell mass is hypothesized to decline and immunologic abnormalities such as autoantibodies directed against insulin and islet cytopiasmic antigens are found. Stimulated insulin secretion is still preserved at this stage. Over a period of years, however, the progressive loss of P cell,s,l.teads to diminished insulin secretion with intravenous glucose tolerance tests (IVGTT) while the individual is still normoglycemic (Stage IV). Overt diabetes (i.e., diabetes onset or clinical manifestation of disease characterized by hyperglycemia) is Stage V, and can develop years later when approximately 90% of pancreatic P cells are destroyed. In Stage V when overt diabetes is first recognized, some residual insulin production remains (as demonstrated by the presence of the connecting peptide of proinsulin, C
peptide, in the serum) but the individual usually requires exogenous insulin for life. Finally, in Stage VI, even the remaining p cells are destroyed and C
peptide can no longer be detected in the circulation.
While the initiating factor(s) and specific sequence of events leading to diabetes, including the relative importance of different cell types and cytokines, are still widely debated, a key role is generally recognized for self-antigen reactive T cells (Miller et al., 1988 [5]; Harada and Makino, 1986 [6];
Koike et al., 1987 [7]; Makino et al., 1986 [8]). In addition to T lymphocytes, insulitis is characterized by macrophages, dendritic cells (Voorbij et al., 1989 [9]) and B cells, which may serve as professional antigen presenting cells (APC). Macrophages may also destroy islet 0 cells themselves by release of cytokines or free radicals (Nomikos et al., 1986 [10]).
Thus, autoimmune diabetes relies upon both cellular migration and immune stimulation of newly resident cells.

WO 94/17828 21553 0 3 p-qH "TCOA /nI dc6 Cell trafficking to inflammatory sites is regulated by accessory molecules LFA-1, MAC-1 and VLA4 (Larson and Springer, 1990 [11]; Hemler et al., 1990 (12)) on the surface of lymphocytes (LFA-1, VLA4) and macrophages (Mac-i, VLA4), and by their counter-ligands ICAM (for LFA-1 and MAC-1), and VCAM (for VLA4) which are unregulated by cytokines on vascular endothelium (Larson and Springer, 1990 [11]; Lobb, 1992 [13];
Osborn, 1990,[14]). In addition, VLA4 binds to an extracellular matrix component, the CS-1 domain of fibronectin (FN) (Wayner et al., 1989 [15]). The relative importance of these pathways, for example, LFA-1 and VLA4 on lymphocytes or MAC-1 and VLA4 on monocytes, in controlling cell migration is still a subject of investigation. In vitro data suggest that the differential use of these pathways appears to depend upon the activation status of both the leukocytes and endothelial cells (Shimizu et al., 1991 [16]). Their ability to control cell migration to inflammatory sites in vivo has been directly demonstrated with monoclonal antibodies (mAbs) to ICAM, MAC-1 or VLA4 inhibiting various animal models of disease (Barton et al., 1989 (17), phorbol ester-induced rabbit lung inflammation; Issekutz and Issekutz, 1991 [18], delayed type hypersensitivity;
Issekutz, 1991 [19], adjuvant-induced arthritis;
Yednock etal., 1992 [20], transfer of experimental allergic encephalomyelitis (EAE); Lobb, 1992 [21], asthma).
ICAM and VCAM are also found on the surface of macrophages and dendritic cells in lymphoid tissues (Dustin et al., 1986 [22]; Rice et al., 1990 [23]; Rice et al., 1991 [24]). Their distribution on these professional APC is consistent with functional data indicating a role for LFA-1 and VLA4 in T cell ~jS5353 - 4 -activation (Shimuzu et al., 1990 [25], Burkly et al., 1991 [26]). However, numerous other receptor-ligand pairs including CD4/ MHC class II and CD8/MHC class I
(Rudd et al., 1989 [27]), CD2/LFA-31% (Moingeon et al., 1989 [287]), CD28/B7 (Harding et al....' 1992 [29]) may also support adhesion or costimulate T cells during T/APC or T/target cell interactions. The specific contributions of these numerous pathways in the development of diabetes is unresolved. Because there are multiple molecular pathways for cell adhesion and T
cell activation, it is not possible to predict whether intervention in one or more of these pathways might affect onset or severity of diabetes disease, and, in particular, which of these pathways are crucial or relevant to the disease process.
Antibodies directed to T cells have been utilized in murine and rat models for spontaneous diabetes and adoptive transfer of diabetes to deplete T cells and thus prevent disease (see, e.g., Harada and Makino, 1986 [6], anti-Thy 1.2; Koike et al., 1987 [7], Miller et al., 1988 [5] and Shizuru et al., 1988 [30], anti-CD4; Barlow and Like, 1992 [31], anti-CD2; Like et al., 1986 [32], anti-CD5 and anti-CD8). In addition, an antibody directed to the complement receptor type 3 (CR3) molecule or MAC-1 on macrophages has been utilized to prevent macrophage and T cell infiltration of pancreatic tissue in a murine adoptive transfer model of disease (Hutchings et al., 1990 [33]). It is unknown whether VLA4 is relevant to insulitis or to the activity of islet-specific cells after localization in the pancreas.
Current treatment protocols suggested for type I
diabetes have included certain immunomodulatory drugs summarized by Federlin and Becker [34] and references cited therein. A long prediabetic period with immunologic abnormalities and progressive p cell destruction suggests it may be possible to halt p cell loss with immune intervention (Castano and Eisenbarth, 1990 [1]).
Suggested agents/protocols have included certain immunomodulatory and immunosuppressive agents:
levamisol, theophyllin, thymic hormones, ciamexone, anti-thymocyte globulin, interferon, nicotinamide, gamma globulin infusion, plasmapheresis or white cell transfusion. Agents such as cyclosporin A and azathioprine which impair T cell activation and T cell development, respectively, have been used in clinical trials (Zielasek et al., 1989 [35]). The most promising results have been achieved with cyclosporin A
(Castano and Eisenbarth, 1990 [1]). Federlin and Becker, 1990 [34] suggest, however, that cyclosporin A
may not be recommended for general or long-term use because of toxic side effects, at least when given in higher doses. Higher doses of cyclosporin, or in combination with other immunosuppressive drugs, or both, have been associated with the development of lymphoma and irreversible kidney damage (Eisenbarth, 1986 [4]; Eisenbarth, 1987 [36]) Additional studies on other suggested agents are necessary to assess safety and efficacy. Even the cyclosporin A studies show that its efficacy in maintaining remission of diabetes is for one year in about 30-60% of new onset diabetes. Within 3 years, however, remissions are almost invariably lost (Castano and Eisenbarth, 1990 [1]). Treatment protocols after onset of disease are particularly problematic, since, for example, at the time diabetes is diagnosed in humans, insulitis has typically progressed already to a loss of more than 80%
of the p cells. Thus, it is possible that cyclosporin A may be preventing further p cell destruction, but so ~1r~5~p3 - 6 -few 0 cells may be present at the onset of the diabetes that they cannot maintain a non-diabetic state over time (Castano and Eisenbarth, 1990 [1]). Suppression of insulitis and/or prevention of disease may be more successful if the treatment could start at an earlier phase, i.e., before disease onset.
There are two major prerequisites in order to develop any preventative tre4tment for diabetes disease: (1) the ability t6 accurately identify the prediabetic individual and (2) the development of safe, specific and effective preventive treatments.
Significant progress has been made in identifying prediabetic individuals, however, much work remains in the development of safe, specific and effective preventive treatments as discussed and reviewed by Eisenbarth and colleagues (see, e.g., Ziegler and Eisenbarth, 1990 [37]; Ziegler et al., 1990 [38];
Ziegler et al., 1990 [39]). It has been possible to identify certain risk factors and at-risk groups for type I diabetes and thus to predict individuals most likely to go on to clinical disease and to estimate the approximate rate of disease onset in these individuals.
The ability to identify individuals with susceptibility to diabetes or to predict type I diabetes in the pre-clinical stage by the combination of genetic (HLA
typing), immunological (islet and insulin autoantibodies) and metabolic (first phase insulin secretion to intravenous glucose preceding the development of hyperglycemia) markers makes the identification and use of prophylactic immunotherapeutic drugs and protocols possible during the evolution of the autoimmune disease process when cell destruction is only partial. To date, there has been little success, however, in treating human diabetes. Generally, because human treatment has been WO 94/17828 21JJ C C3O3 pCT/US94/41456 used only after onset of ~he disease, treatment was followed by a temporary complete or partial remission only in a certain number of patients. Since immunosuppressive mechanisms may prevent insulitis and/or diabetes, there is a need for immunosuppressive components for use in the prediabetic stage. In particular, there is a need for safer and more specifically actin4 compounds, e.g., monoclonal antibodies, whichinhibit entry of effector cells into the pancreas or function of those cell which may have already entered the islets of Langerhans.
It has now been surprisingly discovered that administering an anti-VLA4 antibody significantly reduced the incidence of diabetes, in a rodent model of diabetes disease. The NOD mouse model of diabetes is a well established model directly comparable to human type-I diabetes. Using an adoptively transferred disease experimental protocol, irradiated non-diabetic NOD mice were administered splenocytes from spontaneously diabetic NOD mice for the acute transfer of the disease. These splenocytes were treated with anti-VLA4 antibody before administration and the recipients were also treated for various periods of time after the transfer with anti-VLA4 antibody.

Z1553~3 - s -SIIlrDSARY OF TH$ INVEZPPION
Accordingly, the present invention provides novel methods for the treatment of insulin dependent (type-I) diabetes in a prediabetic. In particular, the present invention provides a method for the prevention of insulin dependent diabetes cpmprising the step of administering to a prediabetic individual an anti-VLA4 antibody, such as antibody HP1/2 or a humanized anti-VLA4 antibody derived from HP1/2. Also contemplated is the use of analogous antibodies, antibody fragments, soluble proteins and small molecules that mimic the action of anti-VLA4 antibodies in the treatment of diabetes. In addition, the present invention provides a method for the treatment of diabetes by administering to a mammal, including a human, with a susceptibility to diabetes an antibody capable of binding to the a4 subunit of VLA4 in an amount effective to provide inhibition of the onset of diabetes. Also contemplated is the use of recombinant and chimeric antibodies, fragments of such antibodies, polypeptides or small molecules capable of binding a4/VLA4. Also contemplated are soluble forms of the natural binding proteins for VLA 4, including soluble VCAM-1, VCAM-1 peptides or VCAM-1 fusion proteins as well as fibronectin, fibronectin having an alternatively spliced non-type III connecting segment and fibronectin peptides containing the amino acid sequence EILDV or a similar conservatively substituted amino acid sequence.
These agents will act by competing with the cell-surface binding protein for VLA4.

- 8a -According to one aspect of the present invention, there is provided use of an antibody against the a4 subunit of very late antigen 4 (VLA-4), or an antigen-binding fragment thereof, for the manufacture of a medicament for prevention of insulin dependent type I diabetes in a prediabetic individual.

According to another aspect of the present invention, there is provided use of an antibody, a recombinant antibody, a chimeric antibody, or antigen-binding fragment of such antibodies that bind to the a4 subunit of very late antigen 4 (VLA4) for the manufacture of a medicament for treatment of insulin dependent type I
diabetes in a mammal.

According to still another aspect of the present invention, there is provided use of a soluble vascular cell adhesion molecule-1 (VCAM-1) polypeptide for the manufacture of a medicament for treatment of insulin dependent type I
diabetes.

According to yet another aspect of the present invention, there is provided a pharmaceutical composition for use in inhibiting onset of insulin dependent type I
diabetes, wherein the composition consists essentially of a monoclonal antibody recognizing the a4 subunit of very late antigen 4 (VLA4) in a pharmaceutically acceptable carrier.

According to a further aspect of the present invention, there is provided use of an antibody against the a4 subunit of very late antigen 4 (VLA-4), or an antigen-binding fragment thereof, for the treatment of insulin dependent type I diabetes.

- sb -According to yet a further aspect of the present invention, there is provided use of a soluble vascular cell adhesion molecule-i (VCAM-1) polypeptide for the treatment of insulin dependent type I diabetes.

WO 94/17828 2155303 prT/US94/01456 - 9 - , BRIEF DEBCRIPTIOli OF 'PHE DRAWINGS
Figure 1 is a graph depicting the effect of anti-VLA4 antibody (Rl-2) and controls on prevention of diabetes after adoptive transfer of spleen cells; the frequency of recipients which became diabetic and day of disease onset are shown for transfer of 2x107 splenocytes from diabetic (D) NOD donors without treatment (closed circles), with a non-specific rat IgG2b treatment (closed triangles), and with Rl-2 anti-VLA4 treatment (closed diamonds), as well as for transfer of splenocytes from nondiabetic (Y) NOD donors (open squares); the splenocytes were transferred with Rl-2 or rat IgG2b or without mAb, and then R1-2 or rat IgG2b was injected every other day through day 12 post transfer (n=8-10 for all groups).
Figure 2 is a graph depicting the effect of anti-VLA4 antibody (Rl-2) and controls on prevention of diabetes after adoptive transfer of spleen cells; the frequency of recipients which became diabetic and day of disease onset are shown for transfer of 3x107 splenocytes from diabetic (D) NOD donors without treatment (closed circles), with a non-specific rat IgG2b treatment (closed triangles), and with Rl-2 anti-VLA4 treatment (closed diamonds), as well as for transfer of splenocytes from nondiabetic (Y) NOD donors (open squares); the splenocytes were transferred with R1-2 or rat IgG2b or without mAb, and then Rl-2 or rat IgG2b was injected every 3.5 days through day 25 post transfer (n=4-5 for all groups).
Figure 3 is a graph depicting the effect of anti-VLA4 antibody (R1-2) and controls on prevention of diabetes after adoptive transfer of spleen cells; the frequency of recipients which became diabetic and day of disease onset are shown for transfer of 2-3x107 splenocytes from diabetic (D) NOD donors without treatment (closed circles), with a non-specific rat IgG2b treatment (closed triangles), and with R1-2 anti-VLA4 treatment (closed diamonds), as well as for transfer of splenocytes from nondiabetic (Y) NOD donors (open squares) or for PBS alone (open circles); the splenocytes were transferred with Rl-2 or rat IgG2b or without mAb, and then R1-2 or'rat IgG2b was injected every 3.5 days through day 25 post transfer (n=5 for all groups).
Figure 4 is a bar graph depicting the effect of anti-VLA4 antibody (Rl-2) and controls on the degree of insulitis after adoptive transfer of spleen cells; the frequency of uninfiltrated islets (Grade 0-I
infiltrate, stipled bar) and infiltrated islets (Grade II-IV insulitis, solid bar) were quantitated and shown after transfer of cells treated with R1-2, rat IgG2b or without mAb, and then Rl-2 or rat IgG2b injected every 3.5 days through day 25 with mice sacrificed when diabetic or on day 26 post-transfer. Pancreatic sections from n=4-5 mice were scored for each experimental group, i.e., Y-Y (non-diabetic donor cells) or D-Y (diabetic donor cells) into non-diabetic (Y) recipients with no mAb treatment, treatment with rat IgG2b or treatment with R1-2.
Figure 5 is a bar graph depicting the effect of anti-VLA4 antibody (R1-2) and controls on the degree of insulitis after adoptive transfer of spleen cells; the frequency of uninfiltrated islets (Grade 0-I
infiltrate, stipled bar) and infiltrated islets (Grade II-IV insulitis, solid bar) were quantitated and shown after transfer of cells treated with Rl-2, rat IgG2b or without mAb, and then Rl-2 or rat IgG2b injected every other day through day 12 post-transfer, then maintained without further mAb injection until sacrificed when diabetic or on day 29 post-transfer. Pancreatic sections from n-4-5 mice were scored for each experimental group, i.e., Y=Y (non-diabetic donor cells) or D-Y (diabetic donor cells) into non-diabetic (Y) recipients with no mAb treatment, treatment with rat IgG2b or treatment with Rl-2.
Figure 6 is a graph depicting the effect of anti-VLA4 antibody (Ri-2) and controls on prevention of diabetes in a spontaneous disease model for diabetes;
the frequency of recipients which beca=e diabetic and day of disease onset are shown for NOD mice without treatment (open squares), with a non-specifia rat IgG2b treatment (closed circles), and with R1-2 anti-VLA4 treatment (closed trianqles); R1-2 or rat IgG2b was injected for 8 weeks in NOD mice twice weekly from week four to week twelve of aqe.
Fiqure 7 is a graph depicting the effect of VCAM
2D-IgG fusion protein and controls on prevention of diabetes after adoptive transfer of spleen cells; the frequency of recipients which became diabetic and day of disease onset are shown for transfer of 2x107 splenocytes from diabetic (D) NOD donors with an irrelevant rat LFA-31g fusion protein treatment (closed squares), and with VCAM 2D-IgG treatment (open circles) or of recipients which recieved PBS alone without cells transferred (closed triangles); the splenocytes were transferred with VCAM 2D-IgG or rat LFA-3Iq, and then VCAM 2D-IqG or rat LFA-31g was injected every other day through day 17 post-transfer (n - 5 for all groups).
Figure 8 is a schematic depicting structure of VCAM 2DIgG fusion protein described in Example 5. VCAM
2D-IgG is a soluble form of the ligand for Vi.h4 (VCAM1) and consists of the two N-terminal domains of VCAM1 fused to the human IgGi heavy chain constant region sequences (Hinges, CK2 and C.3).

~155~p3 - 12 -- DETAILED DESCRIPTION OF THE INVENTION
The invention relates to a treatment including the prevention of insulin delpendent (type I) diabetes.
More particularly, thtinvention relates to the use of antibodies to VLA44ri the treatment of diabetes in a prediabetic individual. The term "prediabetic" is intended to mean an individual at risk for the development of diabetes disease (e.g., genetically predisposed) at any stage in the disease process prior to overt diabetes or diabetes onset. The term "diabetic" is intended to mean an individual with overt hyperglycemia (i.e., fasting blood glucose levels >_ 250 mg/dL). The term "overt diabetes" or "diabetes onset"
is intended to mean a disease state in which the pancreatic islet cells are destroyed and which is manifested clinically by overt hyperglycemia (i.e., fasting blood glucose levels ? 250 mg/dL).
In the first aspect, the invention provides a method of treatment of diabetes comprising the step of administering a composition capable of binding to, including blocking or coating, the VLA4 antigens on the surface of VLA4-positive cells, including lymphocytes and macrophages. For purposes of the invention, the term "binding to VLA4 antigens" is intended to mean reacting with VLA4 antigens on cells and thereby interfering with interactions between VLA4 antigens and either VCAM-1 or fibronectin on the surface of other cells or thereby inducing a change in the function of the VLA4-positive cells. As demonstrated herein, such binding, including blocking or coating, of VLA4 antigens results in a prevention in or protection against the incidence of diabetes. This demonstration utilized a monoclonal antibody against VLA4 as a binding agent which effectively blocked or coated the VLA4 antigens. Those skilled in the art will recognize WO 94/17828 2155303 "^T/US94/01456 that, given this demonstration, any agentthat can bind to, including those that can block or coat, VLA4 antigens can be successfully used in the method of the invention. Thus,,for purposes of the invention, any agent capable of binding to VLA4 antigens on the surface of VLA4-bearing cells and which may effectively block or coat VLA4 antigens, is considered to be an equivalent of the monoclonal antibody used in the examples herein. For example, the invention contemplates as binding equivalents at least peptides, peptide mimetics, carbohydrates and small molecules capable of binding VLA4 antigens on the surface of VLA4-bearing cells.
In a preferred embodiment, the agent that is used in the method of the invention to bind to, including block or coat, cell-surface VLA4 antigens is a monoclonal antibody or antibody derivative. Preferred antibody derivatives for treatment, in particular for human treatment, include humanized recombinant antibodies, chimeric recombinant antibodies, Fab, Fab', F(ab')2 and F(v) antibody fragments, and monomers or dimers of antibody heavy or light chains or intermixtures thereof. Thus, monoclonal antibodies against VLA4 are a preferred binding agent in the method according to the invention.
The technology for producing monoclonal antibodies is well known. Briefly, an immortal cell line (typically myeloma cells) is fused to lymphocytes (typically splenocytes) from a mammal immunized with whole cells expressing a given antigen, e.g., VLA4, and the culture supernatants of the resulting hybridoma cells are screened for antibodies against the antigen.
(See, generally, Kohler et al., 1975 (40)).
Immunization may be accomplished using standard procedures. The unit dose and immunization regimen WO 94/17828 ~ PCT/US94/01456 53~ - 14 -depend on the species of mammal immunized, its immune status, the body weight of the mammal, etc. Typically, the immunized mammals are bled and the serum from each blood sample is assayed for particular antibodies using appropriate screening assays. For example, anti-VLA4 antibodies may be identified by immunoprecipitation of~
125-1abeled cell lysates from VLA4-expressing cells. (See, Sanchez-Madrid et al.
1986 [41] and Hemler et al. 1987 [42]). Anti-VLA4 antibodies may also be identified by flow cytometry, e.g., by measuring fluorescent staining of Ramos cells incubated with an antibody believed to recognize VLA4 (see, Elices et al., (1990) [43]). The lymphocytes used in the production of hybridoma cells typically are isolated from immunized mammals whose sera have already tested positive for the presence of anti-VLA4 antibodies using such screening assays.
Typically, the immortal cell line (e.g., a myeloma cell line) is derived from the same mammalian species as the lymphocytes. Preferred immortal cell lines are mouse myeloma cell lines that are sensitive to culture medium containing hypoxanthine, aminopterin and thymidine ("HAT medium").
Typically, HAT-sensitive mouse myeloma cells are fused to mouse splenocytes using 1500 molecular weight polyethylene glycol ("PEG 1500"). Hybridoma cells resulting from the fusion are then selected using HAT
medium, which kills unfused and unproductively fused myeloma cells (unfused splenocytes die after several days because they are not transformed). Hybridomas producing a desired antibody are detected by screening the hybridoma culture supernatants. For example, hybridomas prepared to produce anti-VI,F,4 antibodies may be screened by testing the hybridoma culture supernatant for secreted antibodies having the ability WO 94/17828 2155303 `,""T/US94/01456 to bind to a recombinant o4-subunit-expressing cell line, such as transfected K-562 cells (see, Elices et al. [43]).
To produce anti-VLA4 antibodies, hybridoma cells that tested positive in such screening assays were cultured in a nutrient medium under conditions and for a time sufficient to,allow the hybridoma cells to secrete the monoclonal antibodies into the culture medium. Tissue culture techniques and culture med"La suitable for hybridoma cells are well known. The conditioned hybridoma culture supernatant may be collected and the anti-VLA4 antibodies optionally further purified by well-known methods.
Alternatively, the desired antibody may be produced by injecting the hybridoma cells into the peritoneal cavity of an unimmunized mouse. The hybridoma cells proliferate in the peritoneal cavity, secreting the antibody which accumulates as ascites fluid. The antibody may be harvested by withdrawing the ascites fluid from the peritoneal cavity with a syringe.
Several mouse anti-VLA4 monoclonal antibodies have been previously described (see, e.g., Sanchez-Madrid et al., 1986 [41]; Hemler et al., 1987 [42]; Pulido et al., 1991 [44]). These anti-VLA4 monoclonal antibodies such as HP1/2 and other anti-VLA4 antibodies (e.g., mAb HP2/1, HP2/4, L25, P4C2, P4G9) capable of recognizing the a chain of VLA4 will be useful in the methods of treatment according to the present invention. Anti-VLA4 antibodies that will recognize the VLA-acy chain epitopes involved in binding to VCAM-1 and fibronectin ligands (i.e., antibodies which can bind to VLA4 at a site involved in ligand recognition and block VCAM-1 and fibronectin binding) are preferred. Such antibodies have been defined as B epitope-specific PCT,'US94/01456 - 16 - 09/02/1994 antibodies (Bl or B2) (see, Pulido et al. (1991) [36]) and are preferred anti-VLA4 antibodies accordina to the present invention. The Rl-2 antibody used as described herein is a B epitope .ype antibodv.
Human monoclonal anti"oodies a2ainst VL A' are another preferred bindinz aQent ~ whicn mav block or coat VLA4 antiQens in the method of the invention. These mav be prepared usina in vitro-pnmed human splenocytes. as described by Boerner et al.. 1991 [45].
.~Iternativelv. thev may be prepared by repertoire clonina as desc ibed bv Persson et al.. 1991 [46] or by Huang and Stollar. 1991 [47]. Another preferred bindina aQent which ma~- block or coat VLA4 antiaens in the method of the invention is a chimeric recombinant antibody havinQ anti-VL A4 specificitv and a human antibodv constant reQion. Yet another prefe:red bindinQ aaent which mav block or coat VLA4 antiQens in the method of the invention is a humanized recombinant antibody having anti-VLA4 specificity. Humanized antibodies mav be prepared. as exemplified in Jones et al.. 1986 [48]; Riechmann. 1988, [49]:
Queen et al..
1989 [50]: and Orlandi et ai.. 1989 [5 1]. Preferred binding aQents includina chimeric recombinant and humanized recombinant antibodies with B epitope specificity have been prepared and are described in the commonlv owned intemational patent appiication. WO
9416094, published on July ? 1, 1994 [52]. The starting material for the preparation of chimeric (mouse V - human C) and humanized anti-VLA4 antibodies mav be a murine monoclonal anti-VLA4 antibody as previouslv described, a monoclonal anti-VLA4 antibody commercially available (e.g., HP2/1. Amac International. Inc., Westbrook, Maine), or a monoclonal anti-VLA4 antibodv prepared in accordance with the teachinQ herein.
For example, the variable regions of ~~~('~~J u~
BGP\ ISiPCVtEPLACI.DOC ANli~~ISf }~--WO 94/17828 _ z 15 5 3 03 rT/US94/01456 the heavy and light chains of the anti-VLA4 antibody HP1/2 have been cloned, sequenced and expressed in combination with constant regions of human immunoglobulin heavy and light chains. Such a chimeric HP1/2 antibody is similar in specificity and potency to the murine HP1/2.antibody, and may be useful in methods of treatment according to the present invention. The HP1/2 V. DNA sequence and its translated amino acid sequences are set forth in SEQ ID NO: 1 and SEQ ID
NO: 2, respectively. The HP1/2 VK DNA sequence and its translated amino acid sequence are set forth in SEQ ID
NO: 3 and SEQ ID NO: 4, respectively. Similarly, humanized recombinant anti-VLA4 antibodies may be useful in these methods. A preferred humanized recombinant anti-VLA4 antibody is an AS/SVMDY antibody, for example, the AS/SVMDY antibody produced by the cell line deposited with the ATCC on November 3, 1992 and given accession no. CRL 11175. The AS/SVMDY humanized antibody is at least equipotent with or perhaps more potent than the murine HP1/2 antibody. The AS V. DNA
sequence and its translated amino acid sequences are set forth in SEQ ID NO: 5 and SEQ ID NO: 6, respectively. The SVMDY VK DNA sequence and its translated amino acid sequence are set forth in SEQ ID
NO: 7 and SEQ ID NO: 8, respectively.
Those skilled in the art will recognize that any of the above-identified antibody or antibody derivative binding agents can also act in the method of the invention by binding to the receptor for VLA4, and may block or coat the cell-surface VLA4 antigen. Thus, antibody and antibody derivative binding agents according to the invention may include embodiments having binding specificity for VCAM-1 or fibronectin, since these molecules appear to either be important in the adhesion cells or the extracellular matrix or . ~ 3 interfere with traffic of cells through tissues and blood.
Alternatively, the binding agents used in the method according to the invention may not be antibodies or antibody derivatives, but rather may be soluble forms of the natural binding proteins for VLA4. These binding agents include soluble VCAM-1, VCAM-1 peptides, or VCAM-1 fusion proteinsAs well as fibronectin, fibronectin having an alternatively spliced non-type III connecting segment and fibronectin peptides containing the amino acid sequence EILDV or a similar conservatively substituted amino acid sequence. These binding agents will act by competing with the cell-surface binding protein for VLA4.
In this method according to the first aspect of the invention, VLA4 binding agents are preferably administered parenterally. The VLA4 binding agents are preferably administered as a sterile pharmaceutical composition containing a pharmaceutically acceptable carrier, which may be any of the numerous well known carriers, such as water, saline, phosphate buffered saline, dextrose, glycerol, ethanol, and the like, or combinations thereof. Preferably, the VLA4 binding agent, if an antibody or antibody derivative, will be administered at a dose ranging between about 0.1 mg/kg body weight/day and about 20 mg/kg body weight/day, preferably ranging between about 0.1 mg/kg body weight/day and about 10 mg/kg body weight/day and at intervals of every 1-14 days. For non-antibody or antibody derivative binding agents, the dose range should preferably be between molar equivalent amounts to these amounts of antibody. Preferably, an antibody composition is administered in an amount effective to provide a plasma level of antibody of at least 1 g/ml.
Optimization of dosages can be determined by 2 155 3 03 rT/US94/01456 administration of the binding agents, followed by assessment of the coating of VLA4-positive cells by the agent over time after administered at a given dose in vivo. Peripheral blood mononuclear cells contained in a sample of the individual's peripheral blood should be probed for the presence of the agent in vitro (or ex vivo) using a second reagent to detect the administered agent. For example, this may be a fluorochrome labelled antibody specific for the administered agent which is then measured by standard FACS (fluorescence activated cell sorter) analysis. Alternatively, presence of the administered agent may be detected in vitro (or ex vivo) by the inability or decreased ability of the individual's cells to bind the same agent which has been itself labelled (e.g., by a fluorochrome). The preferred dosage should produce detectable coating of the vast majority of VLA4-positive cells. Preferably, coating is sustained in the case of a monoclonal antibody or monoclonal antibody derivative for a 1-14 day period.
In practicing this invention, treatment with VLA4 binding agents is preferrably continued for as long as the prediabetic subject maintains a stable normoglycemic plasma level and a stable prediabetic state as reflected by a number of known markers as described above. In the Examples which follow, it has been found that anti-VLA4 mAb, e.g., R1-2 mAb, administration prevented diabetes onset during treatment and that the residual beneficial results of treatment were extended as long as two months following cessation of R1-2 treatment. To sustain the full protective effect of the VLA4 binding agent against diabetes onset, however, continuous treatment with the binding agents is preferred.

WO 94 1j8SW53p3 PCT/US94/01456 The method of the present invention comprises administering to a prediabetic individual a composition comprising an anti-VLA4 antibody. The examples below set forth the results observed in a rodent model of disease. These results deponstrate a protective effect of anti-VLA4 antibpdy in disease onset in the acute transfer model ofrthe disease. The non-obese diabetic (NOD) mouse has become an important model of type I or insulin dependent diabetes mellitus since its introduction by Makino et al., 1980 [7] and has been documented as a particularly relevant model for human diabetes (see, e.g., Castano and Eisenbarth [1], Miller et al., 1988 [5], Hutchings et al., 1990 [33] and references cited therein). That the diabetic syndromes displayed in the NOD mouse and human are similar has been shown by several lines of evidence. For example, in both the NOD mouse and human [1], there is a strong genetic association of diabetes with loci of the major histocompatibility complex. In addition, for example, in both species, an autoimmune pathogenesis is evidenced by (i) the presence of lymphocytic inflammation in the pancreatic islets (i.e., insulitis) that appears to mediate the selective destruction of 0 cells, (ii) the presence of anti-islet cell antibodies, and (iii) the modulating effects of cyclosporin A.
Further evidence in the NOD mouse for an autoimmune etiology of diabetes disease is (i) the ability to transfer diabetes with spleen cells (including purified splenic T cells) from diabetic donors, (ii) prevention of diabetes by in vivo treatment with antibodies specific for T cells, and (iii) failure of a thymic nude mice with NOD genetic background to develop moulitis or diabetes (see, e.g., Miller et al., 1988 [5], Hutchings et al., 1990 [33] and references cited therein).

WO 94/17828 ~ PCT '111456 Although the precise events resulting in diabetes remain unclear, in the NOD mouse a progressive inflammatory response in the pancreas appears to be the initial histological lesion which begins as a periductal /perivascular mononuclear cell infiltrate at 3-4 weeks of age. At about 4-6 weeks of age, insulitis may be observed and beginning at about 12 weeks of age, overt diabetes (i.e., consistent values of 1+ or higher using a Testape (Eli Lilly, Indianapolis, IN) assay for glycosuria or greater than 250 mg/dL if plasma glucose is monitored) occurs. To avoid variations in the immune status of the animals, the NOD mice are obtained from a specific pathogen-free colony and exhibit stable, high incidence of diabetes of about 80% of females and 20% of males which typically become diabetic by about 20 weeks of age. The preferred source for the NOD mice used in the experiments described herein is Taconic Farms (Germantown, NY). A
large body of data, particularly from studies of the BB
rat and NOD mouse has indicated that type I diabetes may be a T-cell mediated disease. Evidence to date suggests an important role for both major T cell subpopulations (CD4/L3T4 and CD8/Ly2) in the development of diabetes in man and in the NOD mouse.
The data supporting the essential role of T cells in diabetes do not exclude the possibility that T
lymphocytes may recruit other cells (e.g., macrophages) as the final effectors for P cell destruction.
Macrophages have been implicated in the disease process based on their presence in the infiltrated islet and the ability of chronic silica treatment to prevent disease (see, e.g., Hutchings et al., 1990 [33] and references cited therein).
Using the NOD strain of mice, investigators have developed an acute transfer model of disease which -~1553~

parallels the spontaneous disease model in that transferred cells derived from diabetogenic NOD mice mediate the disease process, which is characterized by immune reactive cells that mediate insulitis and islet P cell-specific destruction. Moreover, in this model, certain monoclonal antibodies against T cells (see, e.g., Miller et al., 1988 [5Iikr and macrophages (see, e.g., Hutchings et al., 1990 [33] have been shown to abrogate disease onset. Such monocl.onal antibodies have been used in the treatment of spontaneous disease and adoptively transferred disease, for example, anti-CD4 antibody has been shown to abrogate disease in both models (Miller et al, 1988 [5] and Shizuru et al., 1988 [30]). Results of treatment with an agent in the adoptive transfer model or spontaneous disease model are indicative of the ability of the agent to modulate the human disease process.

WO 94/17828 2155303 ^#-T/US94/01456 $ftect of Anti-VLA4 Antibody Treatsent on Adoptive Transfer of Diabetes For the adoptive transfer of diabetes experiments, NOD mice were obtained from Taconic Farms (Germantown, NY) or from the,Joslin Diabetes Center (Boston, MA).
Spontaneously diabetic (D) females of recent onset (13-20 weeks of age) were used as spleen cell donors and 8 week old nondiabetic (Y) females served as recipients.
Spleen cells from 4 week old nondiabetic (Y) female donors which fail to transfer disease were used as a negative control.
Recipient mice were placed on acidified water (1:8400 dilution of concentrated HC1 in water) one week prior to sublethal irradiation (775 rad) performed in split doses (300 rad, 300 rad, and 175 rad) on each of three days (day -2, -1, and the day of transfer), in order to minimize any incidence of intestinal infection subsequent to high dose irradiation (Gamma Cell 1000 Cesium 137 source, Nordion International, Inc., Ontario, Canada). Spleens were harvested from diabetic donors or from nondiabetic controls, cell suspensions made and red cells lysed with Hemolytic Geys solution. Spleen cells were injected intravenously (2-3 x 107 in 0.2 ml PBS) pretreated with either 75 g Rl-2 monoclonal antibody (mAb), 75 g rat IgG2b, or untreated. For the antibody treatment, cells were simply suspended at 1-1.5 X 108 cells/ml with mAb at 375 g/ml and kept on ice until injection. The timing of injection was within 3 hours after last irradiation. Some recipients received PBS alone. The anti-VLA4 mAb Rl-2 and isotype-matched rat IgG2b were purchased from Pharmingen (La Jolla, CA). The R1-2 (rat anti-mouse) anti-VLA4 mAb was originally described by Holzmann et al., 1989 [53]. The Rl-2 anti-VLA4 mAb blocks VLA4 binding to its ligands (Hession et al., 1992 [54]) and therefore belonqs by definition to the B qroup (Pulido at al., 1991 [44], i.e., is equivalent to anti-human VLA4 mAbs of the B group (e.q., HP1/2 or HP2/1).
The Rl-2 mAb or rat IgG2b was administered at a dose of 75 q/0.2 ml intraperitoneally every 2-3 days, a dosinq regimen which was determined to maintain maximal coating of vLA4-positive cells in the peripheral blood, lymphoid organs and bone marrow as detected by staining of peripheral blood cells and single cell suspensions prepared from these organs with a fluorochrome labelled mAb specific for the Rl-2 mAb and FACS analysis to measure fluorochrome positive cells (as described above). Injections were maintained through day 12 or day 24 post transfer. Mice were monitored for diabetes by testing for glycosuria with TesTape (Eli Lilly, Indianapolis, IN) and by plasma glucose levels(Glucometerm, 3 Blood Glucose Meter, Miles, Inc., Elkhart, IN) and were considered diabetic after two consecutive urine positive tests [Testape values.of [+1] or higher] or plasma glucose levels >250 . mg/dL.
An inhibitory effect of the anti-VLA4 mAb on the onset of diabetes was demonstrated when spleen cells isolated from NOD diabetic donors were treated with a saturating quantity of anti-VLA4 mAb R1-2 followed by transfer into nondiabetic irradiated hosts, as described above, and the R1-2 mAb was then administered every other day for 12 days in order to maintain maximal coating of all VLA4-positive cells in the peripheral blood and lymphoid organs for two weeks.
Figure 1 shows the frequency of recipients that became diabetic and the day of disease onset for transfer of 2x107 splenocytes from diabetic NOD donor (D-Y) (i) WO 94/17828 215 5 3 0.1 --T/US94/01456 without treatment (closed circles); (ii) with rat IgG2b treatment (closed triangles), and (iii) with Rl-2 anti-VLA4 treatment (closed diamonds) as well as for transfer of splenocytes from non-diabetic NOD donors (Y-Y) (open squares). Injection of PBS alone gave 0%
incidence. Under these conditions, only 1 of 8 individual R1-2 mAb treated recipients became diabetic, with onset on day 29. post transfer. By contrast, 6/10 and 5/9 individuals became diabetic after receiving splenocytes from diabetic donors treated with no mAb or with non-specific rat IgG2b, respectively. As shown in Figure 1, diabetes onset occurred as early as day 14 post transfer, though administration of the irrelevant rat IgG2b somewhat delayed onset.
These data demonstrate a protective effect of the Ri-2 mAb which was dependent upon its specificity for VLA4. Recipients of splenocytes from nondiabetic mice or of PBS alone failed to become diabetic. Thus, treatment with anti-VLA4 antibody reduced the frequency of diabetes during 30 days post transfer.
Although the results shown in Figure 1 demonstrate that clinical diabetes occurred in only 1 of 8 anti-VLA4 treated animals, it was possible that the anti-VLA4 antibody caused only a minor delay in the onset of disease. Plasma glucose levels were monitored in parallel with urine glucose in order to quantify any increase in blood sugar levels and thereby detect progression to clinical disease. In the anti-VLA4 antibody treated group shown in Figure 1, all mice were still normoglycemic on day 29 with an average plasma glucose value of 100 7 mg/dL, n=7, except for the single individual who scored as clinically diabetic by urine test and plasma glucose >500 mg/dL. Thus, disease progression was not apparent in any of the other anti-VLA4 antibody treated recipients shown in Figure 1 on day 29 post transfer, a full 2 weeks beyond the last anti-VLA4 antibody injection. Analysis of sera from these mice confirmed that the anti-VLA4 mAb dropped to low or undetectable levels by day 18-21 post-transfer.
Additional cell transfers'were performed in order to confirm that the anti-VI+A4 mAb protected against transfer of diabetes. In these experiments, the anti-VLA4 antibody treatment was extended to day 25 post transfer but administered every 3.5 days thereby maintaining saturating levels of Rl-2 mAb or rat IgG2b through day 26 when mice were sacrificed for pancreatic tissue. Under these conditions, an inhibitory effect of the anti-VLA4 mAb on the onset of diabetes was also demonstrated upon spleen cell transfer and R1-2 treatment. Figure 2 shows the frequency of recipients (n=4-5 for each group) that became diabetic and the day of disease onset for transfer of 3x10' splenocytes from diabetic NOD donors (D-Y) (i) without treatment (closed circles), (ii) with IgG2b treatment (closed triangles) and with Ri-2 anti-VLA4 treatment (closed diamonds), as well as for transfer of splenocytes from non-diabetic NOD donors (Y--Y; open squares). Injection of PBS alone gave 0% incidence. Figure 2 shows that only 1 out of 5 Rl-2 mAb treated mice became diabetic by day 22 post transfer whereas diabetes was transferred in 4/4 recipients without R1-2 mAb and 5/5 treated with rat IgG2b. Disease onset occurred as early as day 13 post transfer. These experiments, individually and collectively demonstrate that anti-VLA4 mAb reproducibly protects against development of diabetes in an acute transfer model of disease.
Further experiments were performed to determine whether the anti-VLA-4 mAb simply delayed disease onset during the treatment period or if it could achieve a longer-term protective effect. Figure 3 shows the onset of diabetes in mice over time after Rl-2 injection (once every 3.5 days through day 25) with only 2/5 mice becoming diabetic on days 35 and 38 post transfer, 10-13 days after the last R1-2 injection. By contrast, diabetes occurred in the untreated and IgG2b treated groups as-early as day 11 post transfer, with 100% incidence by days 18-21. Surprisingly, disease incidence in the Rl-2 treated group did not further increase even as long as 2 months following the last Rl-2 injection. Plasma glucose values monitored in parallel during this time reveal that these three individuals were consistently normoglycemic. After this point (i.e., approximately 3 months post-transfer), even the negative control groups which received PBS alone or non-diabetic cells begin developing spontaneous disease. In summary, the VLA-4-specific mAb reduces the incidence of diabetes transfer. Moreover, its protective effect against disease is sustained in the absence of further mAb treatment.

1~53a~ - 28 -$BAMPLE 2 Bffect of Anti-VLA4 aAb on Pancreatis Insulitis For histological analysis, mice were sacrificed between 2-4 weeks post-transfer as described in this Example and pancreata harvested in 10% formalin buffered saline for paraf'f,iri-embedded sections which were stained with hematAxylin and eosin (H&E) for histology. Degree of insulitis was scored as follows:
Grade 0: no insulitis [islet devoid of inflammation];
Grade I: peri-insulitis [inflammatory mononuclear cells located peripheral to the islet]; Grade II: <25%
infiltrated [<25% of the islet interior contains lymphocytic inflammatory cells]; Grade III: 25-50%
infiltrated [lymphocytic infiltration); Grade IV: >50%
infiltrated. The percent of islets in each Grade was then calculated relative to the total number of islets examined. Histologic sections were examined and scored for the degree of insulitis following the adoptive transfer of NOD splenocytes with and without anti-VLA4 mAb treatment and the results tabulated. Specifically, the frequency of uninfiltrated islets (Grade 0-I
infiltrate) and islets with Grade II-IV insulitis (as described above) were quantitated. For each experimental group, pancreatic sections from n= 4-5 mice were scored.
Pancreatic tissue was recovered from recipients treated with the anti-VLA-4 mAb for various time periods in order to address its effect on the establishment of islet-specific cellular infiltrates.
Mice were treated with nonspecific rat IgG2b or Rl-2 mAb every 3.5 days through day 14 when sacrificed.
Similarly, mice were treated through day 25 and sacrificed after diabetes was diagnosed or on day 26 post transfer. Mice continuously treated with the R1-2 mAb for 14 days post transfer maintain a high frequency WO 94/17828 _ 215 5 3 0 3 -"ca4/01456 (76%) of uninfiltrated islets, with only 24%
progressing to grade II-IV insulitis. By contrast those treated with nonspecific rat IgG2b show the reciprocal pattern, with 74% severe insulitis.
Likewise, in the mice treated with Rl-2 though day 25 (20% diabetic, pancreata isolated from mice reported in Figure 2), a high frequency (58%) of uninfiltrated islets were preserved, similar to that (55%
uninfiltrated) in nondiabetic recipients of young NOD
splenocytes, as shown in Figure 4. By contrast, both the untreated or IgG2b-treated mice had only 28%
uninfiltrated islets, and conversely had increased (72%) insulitis. Thus, the anti-VLA-4 mAb treatment appears to specifically inhibit or alternatively to delay the development of insulitis upon adoptive transfer of diabetogenic spleen cells.
In order to distinguish between these alternatives, the pattern of insulitis after 4 weeks post transfer was determined when mice were treated with rat IgG2b or Rl-2 mAb through day 12 and then maintained without further treatment. Mice were sacrificed upon diabetes diagnosis or on day 29 post transfer. Analysis of sera from these mice confirmed that circulating anti-VLA-4 mAb dropped to undetectable levels by days 18-21 post transfer. With this protocol, the degree of insulitis in the Rl-2-treated group (69% insulitis, 25% diabetic) was similar to that in untreated recipients (73% insulitis, 60% diabetic) though still lower than that in the rat IgG2b-treated mice (96% insulitis, 75% diabetic), as shown in Figure 5. Significantly, the severity of insulitis was similar between the Rl-2 treated, untreated and rat IgG2b treated groups with an average of 57%, 47%, 64%
Grade III/IV infiltrates, respectively. Even considering only the nondiabetic Rl-2 treated WO 94/17828 j1553p3 - 30 - PCT/US94/01456 individuals, they still exhibited 59% insulitis with 52% Grade III/IV infiltrates. Recipients of nondiabetogenic NOD splenocytes had only 7% Grade III/IV infiltrates. Conversely, Figure 5 shows that the frequency of uninfiltrated islets was decreased in the Rl-2 treated mice as compared to recipients of saline or nondiabetogenic spleen cells. Thus, the cde ree of insulitis g progret ed in these Rl-2 treated mice (Figure 5) as compared to mice wherein Rl-2 treatment was maintained (Figure 4) and approached that in the untreated and rat IgG2b treated control groups.
Taken together, these data indicate that anti-VLA-4 mAb administration can delay the progression of insulitis in an acute transfer model of disease.

WO 94/17828 21 5 5 3 0 3 1T'US94/01456 BZPiMPLE 3 Comparison of Dilterent Anti-VLA4 ]lntibody Trea_taent on Adoptive Transfer of Diabetes This Example provides comparative efficacy results of PS/2, an anti-VLA4 antibody, with Rl-2 using the adoptive transfer model and procedure described in Example 1. NOD mice were treated with (a) an irrelevant control antibody (D/rat IgG2b, n = 19 mice);
(b) Rl-2 antibody (D/R1-2 mAb, n = 24 mice); (c) PS/2 mAb (D/PS/2 mAb, n = 5 mice); or (d) no treatment (NONE, n = 26 mice). Spleen cells were injected intravenously (2-3x 107 in 0.2 ml PBS) and pretreated with either 75 g Rl-2 mAb, 75 g PS/2 mAb, 75 g rat IgG2b, or untreated. Isolation and purification of PS/2 anti-VLA4 mAb was originally described by Miyake et al., 1991 [55].
The Rl-2 mAb, PS/2 mAb or rat IgG2b was administered at a dose of 75 g/0.2 ml intraperitoneally every 2-3 days, a dosing regimen which was determined to maintain maximal coating of VLA4-positive cells in the peripheral blood, lymphoid organs and bone marrow as detected by staining of peripheral blood cells and single cell suspensions prepared from these organs with a fluorochrome labelled mAb specific for the Rl-2 and PS/2 mAb and FACS
analysis to measure fluorochrome positive cells (as described above). Injections were maintained through days 22 to 25 post transfer. Mice were monitored for diabetes by testing for glycosuria with TesTape (Eli Lilly, Indianapolis, IN) and by plasma glucose levels (Glucometer, 3 Blood Glucose Meter, Miles, Inc., Elkhart, IN) and were considered diabetic after two consecutive urine positive tests [Testape values of [+1] or higher] or plasma glucose levels >250 mg/dL.

An inhibitory effect of the anti-VLA4 mAb on the onset of diabetes was demonstrated when spleen cells isolated from NOD diabetic donors were treated with a saturating quantity of anti-VLA4 mAb Rl-2 or PS/2 followed by transfer into nondiabetic irradiated hosts, as described above, and the Rl-2 mAb or PS/2 mAb was then administered every'other day for 22-25 days in order to maintain maximal coating of all VLA4-positive cells in the peripheral blood and lymphoid organs for about two weeks. Table 1 shows the frequency of recipients that became diabetic and the day of disease onset for transfer of splenocytes from diabetic NOD
donor (i) without treatment (D); (ii) with rat IgG2b treatment (D/nonspecific rat IgG2b); (iii) with Rl-2 anti-VLA4 treatment (D/R1-2 mAb); (iv) with PS/2 treatment (D/PS/2 mAb) as well as for transfer of splenocytes from non-diabetic NOD donors (non-D). Non-diabetic mice receiving PBS and no splenocytes (NONE) were included as a control. Injection of PBS alone gave 4% incidence. Under these conditions, only 1 of 24 individual R1-2 mAb treated recipients became diabetic, with onset on day 22 post transfer while none of the five individual PS/2 mAb treated recipients became diabetic. By contrast, 16/19 individuals became diabetic after receiving splenocytes from diabetic donors treated with no mAb or with non-specific rat IgG2b. As shown in Table 1, diabetes onset occurred as early as day 14 post transfer, though administration of the irrelevant rat IgG2b somewhat delayed onset by one day.
These data demonstrate a protective effect of the Rl-2 mAb and PS/2 which were dependent upon its specificity for VLA4. Recipients of splenocytes from nondiabetic mice or of PBS alone failed to become diabetic. Thus, treatment with anti-VLA4 antibody reduced the frequency of diabetes during 30 days post transfer. Analysis of sera from these mice confirmed that levels of Rl-2 and PS/2 anti-VLA4 mAb become undetectable between days 26 and 34 post-transfer.

Anti-VLA-4 mAbs Inhibit Adoptive Transfer of Diabetes in NOD Mice Cells Transferred/Treataient= No. Diabetic/Total Recipisnts+ Day of Onaet XsSEM

NONE 1/26 (4%) 34 Non-D 1/15 (7%) 15 D 16/19 (84%) 14 0.2 D/Nonspecific rat IgG2b 16/19 (84%) 15 0.9 D/Rl-2 mAb 1/24 (4%) 22 D/PS/2 mAb 0/5 (0%) 'Spleen cells from 4 week old nondiabetic (NON-D) or from new onset diabetic (D) NOD females were transferred, with D cells suspended in mAb or rat IgG
or without mAb before transfer and recipients treated twice weekly for 22-25 days. Mice were monitored for one month post transfer. Data are compiled from 5 experiments.
+D/Rl-2 and D/PS/2 mAb treated groups are significantly different from D and D/rat IgG2b treated groups by Chi square test with Yates' correction as follows: Rl-2 vs. IgG2b treated and D group, p<0.0001;
PS/2 vs. IgG2b treated and D group, p<0.003.

5~3p3 - 34 -Effect of Anti-VLA4 Antibody Treatment on Bpontaneous Diabetws xodal This Example described efficacy results using Rl-2 mAb in the spontaneous diabetes model which employs NOD
mice. NOD mice were treat,ed for 8 weeks with (a) an irrelevant control antibody (NOD/rat IgG2b, n = 10 mice); (b) R1-2 antibody (NOD/R1-2, n = 20 mice); or (c) no treatment (NOD, n = 10 mice) starting at week four to week twelve of age. mAb was administered at a dose of 75 g in 0.2 ml PBS iv, twice weekly. Mice were monitored for diabetic events by TesTape for glycosuria as previously described.
Figure 6 demonstrates a marked delay in diabetes onset (12-16 weeks delay) following R1-2 administration, as compared to the two control groups.
NOD mice which received irrelevant IgG2b mAb or no treatment developed diabetes as early as 13 weeks.
These spontaneous disease model results parallel the adoptive transfer results with Rl-2 mAb illustrated in Figure 1 and directly demonstrate that an anti-V1A4 antibody protects against diabetes onset.

WO 94/17828 2155303 "rTIUS94/01456 ' - 35 -EZA1rIPLE 5 Btfect of a vC7K-Ig tusion Protein on Adoptive Transfer of Diabetes The adoptive transfer experiment described in Example 1 was repeated with a VCAM-Ig fusion protein (VCAM 2D-IgG) instead of an anti-VLA4 mAb. VCAt+i 2D-IgG
is a soluble foriq of the ligand for VLA4 (VCAM1) which consists of the twp;,N-terminal domains of VCAM1 fused to the human IgGl heavy chain constant region sequences (Hinges, CH2 and CN3). The VCAM 2D-IgG DNA sequence and its translated amino acid sequence are shown in SEQ
ID NO: 9. Figure 8 illustrates the fusion protein structure. The fusion protein was constructed by recombinant techniques as described below.

Isolation of cDNA of Human IgGl Heavy Chain Region and Construction of Plasmid DSAB144 In order to isolate a cDNA copy of the human IgGi heavy chain region, RNA was prepared from COS7 cells which has been transiently transfected by the plasmid VCAM1-IgGl (also known as pSAB133). Construction of plasmid VCAM1-IqGl is described in PCT patent application WO 90/13300. The RNA was reverse transcribed to generate cDNA using reverse transcriptase and random hexamers as the primers.
After 30 min. at 42'C, the reverse transcriptase reaction was terminated by incubation of the reaction at 95'C for 5 min. The cDNA was then amplified by PCR
(Polymerase Chain Reaction, see, e.g., Sambrook et al., Molecular Cloning. Vol. 3, pp. 14.1-14.35 (Cold Spring Harbor; 1989)) using the following kinased primers:
7~ 0-31 (SEQ ID NO: 10):

5'TCGTC GAC AAA ACT CAC ACA TGC C
Asp Lys Thr His Thr Cys which contains a UII site, and 15~3~3 - 36 -370-32 (SEQ ID NO: 11):

5' GTAAATGAGT GCGGCGGCCG CCAA, which encodes the carboxy terminal lysine of the IgGl heavy chain constant region, followed by a H2tI site.
The PCR amplified cDNA was purified by agarose gel electrophoresis and glass,bead elution for cloning in plasmid pNN03. Plasmid pNN03 was constructed by removing the synthetic polylinker sequence from the commercially available plasmid pUC8 (Pharmacia, Piscataway, New Jersey) by restriction endonuclease digestion and replacing the synthetic polylinker sequence with the following novel synthetic sequence (SEQ ID NO: 12):
GCGGCCGCGG TCCAACCACC AATCTCAAAG CTTGGTACCC GGGAATTCAG
ATCTGCAGCA TGCTCGAGCT CTAGATATCG ATTCCATGGA TCCTCACATC
CCAATCCGCG GCCGC.
The purified PCR amplified cDNA fragment was ligated to pNN03 which had been cleaved with roRV, dephosphorylated, and purified by low melt agarose gel electrophoresis. The ligation reaction was used to transform E.coli JA221 and the resulting colonies were screened for a plasmid containing an insert of approximately 700 bp. The identity of the correct insert was confirmed by DNA sequence analysis, and the plasmid was designated pSAB144.
Construction of Plasmid pSAB142 The plasmid pSAB142 was constructed as follows.
CDNA prepared from COS cells transfected with pSAB133 (as described in the previous section) was subjected to PCR amplification using obligonucleotides 370-01 and 370-29. Oligonucleotide 370-01 includes a NotI site and the nucleotides corresponding to amino acids 1 through 7 of the VCAM-1 signal sequence WO 94/17828 215 5 3 4 3 rT/US94/01456 (SEQ ID NO: 13):
5' GAGCTCGAGGCGGCCGCACCATGCCTGGGAAGATGGTCGTG
MetProGlyLysMetVaiVal Oligonucleotide 370-29 corresponds to the VCAM-1 amino acids 214-219 and includes a,SAII site (SEQ ID NO: 14):
VAA GTC GAC'TTG CAA TTC TTT TAC
The amplified DNA fragment was ligated to the vector fragment of pNN03, cleaved by = RV.

Construction of gSAB132 pJOD-S (Barsoum, J., DNA and Cell Biol., pp.293-300 (1990)) was modified to insert a unique NotI
site downstream from the adenovirus major late promoter so that NotI fragments could be inserted into the expression vector. pJOD-S was linearized by NotI
cleavage of the plasmid DNA. The protruding 5' termini were blunt-ended using Mung Bean nuclease, and the linearized DNA fragment was purified by low melting temperature agarose gel electrophoresis. The DNA
fragment was religated using T4 DNA ligase. The ligated molecules were then transformed into .E cQli JA221. Colonies were screened for the absence of a = I site. The resulting vector was designated pJOD-S
delta Notl. pJOD-8 delta Notl was linearized using SUI and the 5' termini were dephosphorylated using calf alkaline phosphatase. The linearized DNA fragment was purified by low melting temperature agarose gel eletrophoresis and ligated in the presence of phosphorylated oligonucleotide ACE175, which has the following sequence (SEQ ID NO:15):
TCGACGCGGC CGCG
The ligation mixture was transformed into E.coli JA221, and colonies were screened for the presence of a plasmid having aNotI site. The desired plasmid was named pMDR901.
In order to delete the two SV40 enhancer repeats in the Sv40 promoter which controls transcription of the DHFR cDNA, pMDR901'and pJODAe-tPA (Barsoum, DNA and Cell Biol., 9, pp. 293-300 (1990)), both were cleaved with $tII and =III. The 2578 bp hatII-=III
fragment from pMDR901 and the 5424 bp &-LtII-=III
fragment from pJODse-tPA were isolated by low melting temperature agarose gel electrophoresis and ligated together. Following transformation into E.coli JA221, the resulting plasmid, pMDR902, was isolated. pSAB132 was then formed by eliminating the coRI-NotI fragment of pMDR902 containing the adenovirus major late promoter and replacing it with an 839 bp coRI-No I
fragment from plasmid pCMV-B (Clontech, Palo Alto, California) containing the human cytomegalovirus immediate early promoter and enhancer.

Construction of pSAB146 pSAB144 was cleaved with Sa I and NotI, and the 693 bp fragment isolated. pSAB142 was cleaved with NotI and _$_sUI and the 664 bp fragment was isolated.
The two fragments were ligated to pSAB132 which had been cleaved with IotI, and the 5' termini dephosphorylated by calf alkaline phosphatase. The resulting plasmid, pSAB146, contained the DNA sequence encoding the VCAM-1 signal sequence, the amino terminal 219 amino acids of mature VCAM-1, ten amino acids of the hinge region of IgGl and the CH2 and CH3 constant domains of IgGl.

Production of VCAM 2D-IcrG from a stably transformed CHO cell line A recombinant VCAIM 2D-IgG expression vector was constructed as described below and transfected into CHO
cells to produce a cell line continuously secreting VCAIM 2D-IgG.
The 1.357 kb = I fragment containing the VCAM 2D-IgG coding sequence of pSAB146 was purified by agarose gel electrophoresis. This fragment was ligated into the NotI cloning site of the expression vector pMDR901, which uses the adenovirus 2 major late promoter for heterologous gene expression and the selectable, amplifiable dihydrofolate reductase (dhfr) marker. The ligated DNA was used to transform E.COli DH5. Colonies containing the plasmid with the desired, correctly oriented insert were identified by the presence of 5853 and 3734 bp fragments upon digestion with Hind III; and 4301, 2555, 2293, and 438 bp fragments upon digestion with DgIII. The resultant recombinanat VCAM 2D-IgG
expression vector was designated pEAG100. The identity of the correct insert was confirmed by DNA sequence analysis.
The recombinant expression plasmid pEAG100 was electroporated into dhfr-deficient CHO cells according to the published protocol of J. Barsoum (DNA Cell Biol 9: 293-300, 1990), with the following changes: 200 g of ZMI-linearized pEAG100 plasmid and 200 g of sonicated salmon sperm DNA were used in the electroporation protocol. In addition, cells were selected in alpha-complete medium supplemented with 200 nM methotrexate.
To determine expression levels of secreted VCAM
2D-IgG, clones were transferred to a flat bottom 96 well microtiter plate, grown to confluency and assayed by ELISA as described below.

Wells of Immulon' 2 plates (Dynatech, Chantilly, virqinia) were each coated with anti-VC&i[ Mhb 4*9 (isolated and purified on Protein A Sepharose' as described by Carlos et al, 1990 [56]) with 100 1 of anti-VCAM 4B9 MAb diluted to 10 q/al in 0.05 M sodium carbonate/bicarbonate buffer, pH 9.6, covered with Parafilm, and incubated overniqht at 40C. The next day, the plate contents were dumped out and blocked with 200 1/well of a block buffer (54 fetal calf serum in ix PBS), which had been filtered through a 2 filter. The buffer was removed after a 1 hour .incubation at room temperature and the plates were washed twice with a solution of 0.054 Tween-20 in 1X
PBS. Conditioned medium was added at various dilutions. As a positive control, an anti-mouse Ig was also included. Block buffer and LFA-3TIP constituted as neqative controls. The samples and controls were incubated at room temperature for 2 hours.
The plates were then washed twice with a solution of 0.05$ TVaeen-20'w in 1X PBS. Each well, except for the positive control well, was then filled with 50 l of a 1:2000 dilution of HRP-Donkey anti-human IgG (H+L) (Jackson Immune Research Laboratories, Inc.; West Grove, Pennsylvania) in block buffer. The positive control well was filled with 50 l of a 1:2000 dilution of HRP-Goat anti-mouse lgG (H+L)(Jackson Immune Research Laboratories, Inc.; West Grove, Pennsylvania), in block buffer. The plates were then incubated for 1 hour at room temperature.
The HRP conjugated Ab solutions were removed, and the wells were washed twice with 0.05% Tween-20 in 1X
-PBS. Then, 100 l of HRP-substrate buffer was added to each well at room temperature. HtP-substrate buffer was prepared as follows: 0.5 ml of 42mM 3,31, 5,50-tetramethylbenzidine (TMB), (ICN Immunobioloqicals, WO 94/17828 _ 215 5 3 0 3 -*c94/01456 Lisle, South Carolina, Catalogue No. 980501) in DMSO
(Aldrich) was slowly added to 50 ml of substrate buffer (0.1 M sodium acetate/citric acid, pH4.9); followed by addition of 7.5 l of 30% hydrogen peroxide (Sigma, Catalogue No. H-1009).
The development of a blue color in each well was monitored at 650nm on a microtiter plate reader. After 7-10 minutes, the development was stopped by the addition of 100 l of 2N Sulfuric acid. The resulting yellow color was read at 450nm on a microtiter plate reader. A negative control well was used to blank the machine.

Purification of VCAM 2D-IgG
CHO cells expressing VCAN 2D-IgG were grown in roller bottles on collagen beads. Conditioned medium (5 Liters) was concentrated to 500 ml using an Amicon S1Y10 spiral ultrafiltration cartridge (Amicon, Danvers, MA). The concentrate was diluted with 1 liter of Pierce Protein A binding buffer (Pierce, Rockford, IL) and gravity loaded onto a 10 ml Protein A column (Sepharose 4 Fast Flow, Pharmacia, Piscataway, NJ).
The column was washed 9 times with 10 ml of Protein A
binding buffer and then 7 times with 10 ml of PBS.
VCAM 2D-IgG was eluted with twelve-5 ml steps containing 25 mM H3P04 pH2.8, 100 mM NaCl. The eluted samples were neutralized by adding 0.5 M NazHPOy pH8.6 to 25 mM. Fractions were analyzed for absorbance at 280 nm and by SDS-PAGE. The three peaks fractions of highest purity were pooled, filtered, aliquoted and stored at -70"C. By SDS-PAGE, the product was greater than 95% pure. The material contained less than 1 endotoxin unit per mg of protein. In some instances, it was necessary to further purify the Protein A eluate product on Q-Sepharose FF (Pharmacia). The protein A

WO 94/17828 w n~ PCT/US94/01456 eluate was diluted with 3 volumes of 25 m.*i Tris HC1 pH
8.0 and loaded onto a Q-Sepharose FF column at 10 mg VCAM 2D-IgG per ml of resUn. The VCAM 2D-IgG was then eluted from the Q- Sep"~:ose with PBS.
,=
Evaluation of VCAM 2D-IaG
Spleen cell suspensions were prepared from diabetic donors or from nondiabetic controls as described above. Spleen cells were injected intraveneously (2-3 x 107 in 0.2 ml PBS) and were pretreated with either 100 g VCAM 2D-IgG or 100 g of irrelevant LFA-31g fusion protein control. Another group received PBS alone without cells transferred.
The fusion protein LFA-31g (LFA-3TIP) was isolated and purified as described in PCT US92/02050 and Miller et al., 1993 [57]. The VCAM 2D-IgG fusion protein or irrelevant LFA-31g protein was administered at a dose of 100 g/0.2 ml intraperitoneally twice weekly through day 17. This concentration was sufficient to provide a serum level of fusion protein sufficient to saturate VLA4-positive cells, the serum levels determined by ELISA as described above. Diabetes onset was monitored as described above.
The results of the evaluation are shown in Figure 7. As shown in this Figure, VCAM 2D-IgG fusion protein significantly inhibits the onset of diabetes in recipients of cells from diabetic donor mice (D/VCAM-Ig, open circles) with 60% incidence by day 30 post-transfer, as compared to the mice which received cells from diabetic donor (data not shown) and LFA-31g irrelevant control Ig fusion protein (D/LFA-3 Ig) which had already achieved 60% incidence by day 15 post-transfer. Mice which received no cells (PBS only) did not develop disease. There were n = 5 mice per experimental group.

WO 94/17828 _ 215" 3O 3 PrT/US94/01456 In summary, VIA4 binding agents such as anti-VLA4 antibodies were protective against diabetes disease onset (Examples 1, 3 and 4) and were effective in delaying the progression of insulitirs (Example 2) using a murine model for human diabetes. Other VLA4 binding agents such as soluble-VCAIM derivatives (VCAM 2D-IgG) were also useful in protecting against diabetes disease onset (Example 5). The foregoing examples are intended as an illustration of the method of the present invention and are not presented as a limitation of the invention as claimed hereinafter. From the foregoing disclosure, numerous modifications and additional embodiments of the invention will be apparent to those experienced in this art. For example, actual dosage used, the type of antibody or antibody fragment used, mode of administration, exact composition, time and manner of administration of the treatment, and many other features all may be varied without departing from the description above. All such modifications and additional embodiments are within the contemplation of this application and within the scope of the appended claims.

LIST OF R]lFERENCEB CITED

' [1) Castano and Eisenbarth, 1990, Annu. Rev.
Immunology $: 647-79, "Type I Diabetes:
A Chronic Autoimmune Disease of Human, Mouse,E 4nd Rat"
~.~
,=
(2) Fujita et al., 1982, Biomed. Res. 2:
429-436, "Lymphocytic Insulitis in a Non-obese Diabetic (NOD) Strain of Mice:
An Immunohistochemical and Electron Microscope Investigation"

[3] Foulis et al., 1986, Diabetologia 22:
267, "The histopathology of the pancreas in Type I (insulin-dependent) diabetes mellitus: a 25-year review of deaths in patients under 20 years of age in the United Kingdom"

[4] Eisenbarth, 1986, New Engl. J. Med.
314:1360-1368, "Type I Diabetes Mellitus - A Chronic Autoimmune Disease"

[5] Miller et al., 1988, J. Immunol. 140:
52-58, "Both the Lyt-2+ and L3T4+ T Cell subsets are required for the transfer of diabetes in Nonobese diabetic mice"

[6] Harada and Makino, 1986, Exp. Anim. 35:
501, "Suppression of overt diabetes in NOD mice by Anti-thymocyte serum or anti-Thy 1.2 antibody"

[7) Koike et al., 1987, Diabetes 36: 539, "Preventive effect of monoclonal anti-L3T4 antibody on development of diabetes in NOD mice"

(8] Makino et al., 1986, Exp. Anim. 35: 495, "Absence of insulitis and overt diabetes in athymic nude mice with NOD genetic background"

[9] Voorbij et al., 1989, Diabetes 35: 1623-1629, "Dendritic cells and scavenger macrophages in pancreatic islets of prediabetic BB rats"

[10] Nomikos et al., 1986, Diabetes 35:
11302-1304, "Combined treatment with nicotinamide and desferrioxamine prevents islet allograft destruction in NOD mice"

[11] Larson and Springer, 1990, Immunol. Rev.
~: 181-217, "Structure and Function of Leuj~ccyte Integrins"

[12] Hemler=et al., 1990, Immunol. Rev. IJA:
45-66, "Structure of the Integrin VLA-4 and its Cell-Cell and Cell-matrix adhesion functions"

[13] Lobb, RJR., 1992, Adhesion: Its Role in Inflammatory Diseases. ed. J.M. Harlan and D.Y. Liu, New York: W. H. Freeman.
1-18.

[14] Osborn, L., 1990, Cell Aa: 3-6, "Leukocyte Adhesion to Endothelium in Inflammation"

[15] Wayner et al., 1989, J. Cell. Biol. J.U:
1321-1330, "Identification and Characterization of the T Lymphocyte Adhesion Receptor for an Alternative Cell Attachment Domain (CS-1) in Plasma Fibronectin"

[16] Shimizu et al., 1991, J. Cell Biol. JU:
1203, "Four molecular pathways to T cell adhesion to endothelial cells: roles of LFA-1 VCAM-1 and ELAM-1 and changes in pathway hierarchy under different activation conditions"

[17] Barton et al., 1989, J. Immunol. 143:
1278, "The effect of anti-intercellular adhesion molecule-1 on phorbolester-induced rabbit lung inflammation"

[18] Issekutz, T.B. and Issekutz, A.C., 1991, Clinical Immunol. and Immunopathol. l$:
300-312, "T lymphocyte migration to arthritis joints and dermal inflammation in the rat: differing migration patterns and the involvement of VLA-4"

[19] Issekutz, T.G., 1991, J. Immunol 147:
4178-4184, "Inhibition of In Vivo Lymphocyte Migration to Inflammation and Homing to Lymphoid Tissues by the TA-2 = ~1~~~~ - 46 -Monoclonal Antibody - A Likely Role for VLA-4 In Vivo"

[20] Yednock, et al.., 1992, Nature 356: 63-66, "Preventiotk of experimental autoimmune',.encephalomyelitis by antibodie'~,dgainst a4p1 integrin"
~;,, `

[21] Lobb, U.S. Patent Application Serial No.
07/821,768 filed January 13, 1992, "Treatment for Asthma"

[22] Dustin et al., 1986, J. Immunol. 137:
245-254 "Induction by IL-i and Interferon-y Tissue Distribution, Biochemistry, and Function of a Natural Adherence Molecule (ICAM-1)"

[23] Rice et al., 1990, J. Exp. Med. 1,U:
1369, "Inducible Cell Adhesion Molecule 110 (INCAM-110) Is An Endothelial Receptor for Lymphocytes - A CD11/CD18-Independent Adhesion Mechanism"

[24] Rice et al., 1991, Am. J. Path. Il$:
385393, "Vascular and Nonvascular Expression of INCAM-110"

[25] Shimuzu et al., 1990, Immunol. Rev. 114:
109-144, "Roles of Adhesion Molecules in T-cell recognition: Fundamental Similarities between four integrins on resting human T cells (LFA-1, VLA-4, VLA-5, VLA-6) in expression, binding and costimulation"

[26] Burkly et al., 1991, Eur. J. Immunol.
11: 2871-2875, "Signaling by vascular cell adhesion molecule-1 (VCAM-1) through VLA4 promotes CD3-dependent T
cell proliferation"

[27] Rudd et al., 1989, Immunol. Rev. JU:
225-266, "Molecular Interactions, T-Cell Subsets, and a Role of the CD4/CD8:p561ck Complex in Human T-Cell Activation"

[28] Moingeon et al., 1989, Immunol. Rev.
16U: 111-144, "The Structural Biology of CD2"

WO 94/17828 215 5 3 0 3 'T[US94/01456 [29] Harding et al., 1992, Nature 356: 607-609, "CD28-mediated signalling co-stimulates murine T cells and prevents induction of energy in T cell clones"

[30] Shizuru et al., 1988, Science ZgQ: 659-662, "Immunotherapy of the Nonobese Diabetic Mouse: Treatment with an Antibody to T-Helper Lymphocytes"

[31] Barlow and Like, 1992, Amer. J. Pathol.
111: 1043-1051, "Anti-CD2 Monoclonal Antibodies Prevent Spontaneous and Adoptive Transfer of Diabetes in the BB/Wor Rat"

[32] Like et al., 1986, J. Exp. Med. 164:
1145-1159, "Prevention of Diabetes in Biobreeding/Worchester Rats with Monoclonal Antibodies that Recognize T
Lymphocytes or Natural Killer Cells"

[33] Hutchings et al., 1990, Nature ' 348: 639-642, "Transfer of diabetes in mice prevented by blockade of adhesion-promoting receptor on macrophages"

[34] Federlin and Becker, 1990, Klin.
Wochenschr. J: Suppl. XXI 38-43, "Specific Therapeutic Attempts in Experimental and Clinical Type-I
Diabetes"

[35] Zielasek et al., 1989, Clin. Immunol.
Immunopathol. 52: 347-365, "The Potentially Simple Mathematics of Type I
Diabetes"

[36] Eisenbarth, 1987, Hosp. Prac. U:167-183, "Type I Diabetes: Clinical Implication of Autoimmunity"

[37] Ziegler and Eisenbarth, 1990, Horm. Res.
,U: 144-150, "Multiple Target Antigens in Pre-Type I Diabetes: Implications for Prediction"

[38] Ziegler et al., 1990, Diabetes Care 13:
762-765, "Predicting Type I Diabetes"

[39] Ziegler et al., 1990, J. Autoimmun. 3 Suppi. 1: 69-74, "Type I Diabetes:

-polygenic inheritance, multiple autoantigens and `dual' parameter prediction"

[40] Kohler, G. and Milstein, 1975, C. Nature 21&: 295-497, "Continuous Cultures of Fused Cells Secreting Antibody of Pred1fined Specificity"

(41) Sanchez-Madrid et al., 1986, Eur. J.
Immunol., ]¾: 1343-1349, "VLA-3: A novel polypeptide association within the VLA
molecular complex: cell distribution and biochemical characterization"

[42] Hemler et al., 1987, J. Biol. Chem. 262:
11478-11485, "Characterization of the cell surface heterodimer VLA4 and related peptides"

[43] Elices et al., 1990, Cell 60: 577-584, "VCAM-1 on Activated Endothelium Interacts with the Leukocyte Integrin VLA4 at a Site Distinct from the VLA4/Fibronectin Binding Site"

[44] Pulido et al., 1991, J. Biol. Chem., 266(16): 10241-10245, "Functional Evidence for Three Distinct and Independently Inhibitable Adhesion Activities Mediated by the Human Integrin VLA-4"

[45] Boerner et al., 1991, J. Immunol.
147:86-95, "Production of Antigen-specific Human Monoclonal Antibodies from In Vitro-Primed Human Splenocytes"

[46] Persson et al., 1991, Proc. Natl. Acad.
Sci. USA BA: 2432-2436, "Generation of diverse high-affinity human monoclonal antibodies by repertoire cloning"

[47] Huang and Stollar, 1991, J. Immunol.
Methods 141: 227-236, "Construction of representative immunoglobulin variable region cDNA libraries from human peripheral blood lymphocytes without in vitro stimulation"

[48] Jones et al., 1986, Nature In: 522-525, "Replacing the complementarity-2155303 PCT/US94/014-56 - 49 - 09/02/ 1'994 determininz reQions in a human antibodv with those from a mouse"

[-?9] Rlecnmann. 1988. ~iarsre 332: 323-327.
"ReshaoinQ human antibodies for therapv"
[~0] Queen et al.. 1989. Proc.'Natl. Acad.
Sci. USA 86: i 0029. "A humanized antibodv that bincis to the interleukin 2 receptor"

[~ 11 Oriandi et al.. 1989. Proc. Nati. Acad.
Scl. USA 86:3833 "Clonina immunoQlobulin variable domains for expression by the 1 ~ polvmerase chain reaction"

[~2] International Patent Application No.
WO 9416094, published on Juiti- 21. 1994.
"Recombinant Anti-VLA4 Antibodv Molecules"
[~3] Holzmann et al. 1989, Cell 56: 37-46.
"Identification of a Murine Pever's Patch-Specific Lvmphocyte Homin2 Receptor as an InteQrin Molecule with a Chain Homoloeous to Human VLA-4a"

[54] Hession et al., 1992, Biochem. Biophvs.
Res. Commun. 1$3: 163-169, "Cloning of Murine and Rat Vascular Cell Adhesion '0 Molecule-1"
[~ ~] Mivake et al., 1991. J. Exp. Med. 1?~: 599-607.

35 [56] Carlos et al.. 1990, Blood 17 : 965, "Vascular Cell Adhesion molecule-1 (VCAM-1) mediates Lvmphocvte Adherence to Cvtol:ine-activated Cultured Human Endothelial Cells."
~
[57] Miller et al.. 1993, J. Exp. Med. 178.
211.

acP\1 5 1 FCViEPLAC2.DOC JfiL
L~

WO 94/17828 215 5 3 0 3 0rT/US94/01456 The foregoing documents are incorporated herein by reference in their entirety.

SEQUENCE LISTING

(1) GENERAL INFORMATION:
(i) APPLICANT
(A) NAME: Biogen, Inc.
(B) STREET: Fourteen Gambridge Center (C) CITY: Cambridge (D) STATE: Massachusetts (E) COUNTRY: USA
(F) POSTAL CODE (ZIP): 02142 (G) TELEPHONE: 617-252-9200 (H) TELEFAX: 617-252-9595 (ii) TITLE OF INVENTION: Treatment for Insulin Dependent Diabetes (iii) NUMBER OF SEQUENCES: 15 (iv) COMPUTER READABLE FORM:
(A) MEDIUM TYPE: Floppy disk (B) COMPUTER: IBM PC compatible (C) OPERATING SYSTEM: PC-DOS/MS-DOS
(D) SOFTWARE: Patentln Release #1.0, Version #1.25 (EPO) (v) CURRENT APPLICATION DATA:
APPLICATION NUMBER: PCT/US94/01456 (2) INFORMATION FOR SEQ ID N0:1:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 360 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA
(ix) FEATURE:
(A) NAME/KEY: misc_feature (B) LOCATION: 1 (D) OTHER INFORMATION: /notea "pBAG159 insert: HP1/2 heavy chain variable region; amino acid 1 is Glu (E) but Gln (Q) may be substituted"

(ix) FEATURE:
(A) NAME/KEY: CDS
(B) LOCATION: 1-360 (xi) SEQUENCE DESCRIPTION: SEQ ID N0:1:

~;DED SHEET
WE"

Val Lys Leu Gln Gln Ser Gly Ala Glu Leu Val Lys Pro Gly Ala Ser Val Lys Leu Ser Cys Thr Ala Ser Gly Phe Asn Ile Lys Asp Thr Tyr Met His Trp Val Lys Gln Arg Pro Glu Gin Gly Leu Glu Trp Ile Gly Arg Ile Asp Pro Ala Ser Gly Asp Thr Lys Tyr Asp Pro Lys Phe Gln Val Lys Ala Thr Ile Thr Ala Asp Thr Ser Ser Asn Thr Ala Trp Leu Gln Leu Ser Ser Leu Thr Ser Glu Asp Thr Ala Val Tyr Tyr Cys Ala Asp Gly Met Trp Val Ser Thr Gly Tyr Ala Leu Asp Phe Trp Gly Gln Gly Thr Thr Val Thr Val Ser Ser (2) INFORMATION FOR SEQ ID NO:2:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 120 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:

Val Lys Leu Gln Gln Ser Gly Ala Glu Leu Val Lys Pro Gly Ala Ser Val Lys Leu Ser Cys Thr Ala Ser Gly Phe Asn Ile Lys Asp Thr Tyr Met His Trp Val Lys Gln Arg Pro Glu Gln Gly Leu Glu AMENDED S!"EE i Trp Ile Gly Arg Ile Asp Pro Ala Ser Gly Asp Thr Lys Tyr Asp Pro Lys Phe Gin Val Lys Ala Thr Ile Thr Ala Asp Thr Ser Ser Asn Thr Ala Trp Leu Gln Leu Ser Ser Leu Thr Ser Glu Asp Thr Ala Val Tyr Tyr Cys Ala Asp Gly Met Trp Val Ser Thr Gly Tyr Ala Leu Asp Phe Trp Gly Gln Gly Thr Thr Val Thr Val Ser Ser (2) INFORMATION FOR SEQ ID NO:3:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 318 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: double (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA

(ix) FEATURE:
(A) NAME/KEY: CDS
(B) LOCATION: 1-318 (D) OTHER INFORMATION: /product- "HP1/2 light chain variable region"

(ix) FEATURE:
(A) NAME/KEY: misc-feature (B) LOCATION: 1 (D) OTHER INFORMATION: /note= "pBAG172 insert: HP1/2 light chain variable region"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:3:

Ser Ile Val Met Thr Gln Thr Pro Lys Phe Leu Leu Val Ser Ala Gly Asp Arg Val Thr Ile Thr Cys Lys Ala Ser G1n Ser Val Thr Asn Asp Val Ala Trp Tyr G1n Gln Lys Pro Gly Gln Ser Pro Lys 35 .40 45 AMENDED SHEET

215~303 Leu Leu Ile Tyr Tyr Ala Ser Asn Arg Tyr Thr Gly Val Pro Asp Arg Phe Thr Gly Ser Gly Tyr Gly Thr Asp Phe Thr Phe Thr Ile Ser Thr Val Gln Ala Glu Asp Leu Ala Val Tyr Phe Cys Gln Gln Asp Tyr Ser Ser Pro Tyr Thr Phe Gly Gly Gly Thr Lys Leu Glu Ile (2) INFORMATION FOR SEQ ID NO:4:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 106 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO:4:

Ser Ile Val Met Thr Gln Thr Pro Lys Phe Leu Leu Val Ser Ala Gly Asp Arg Val Thr Ile Thr Cys Lys Ala Ser Gln Ser Val Thr Asn Asp Val Ala Trp Tyr Gln Gln Lys Pro Gly Gln Ser Pro Lys Leu Leu Ile Tyr Tyr Ala Ser Asn Arg Tyr Thr Gly Val Pro Asp Arg Phe Thr Gly Ser Gly Tyr Gly Thr Asp Phe Thr Phe Thr Ile Ser Thr Val Gin Ala Glu Asp Leu Ala Val Tyr Phe Cys Gln Gln Asp Tyr Ser Ser Pro Tyr Thr Phe Gly Gly Gly Thr Lys Leu Glu Ile AI ' '"= E T

(2) INFORMATION FOR SEQ ID NO:5:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 429 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA
(ix) FEATURE:
(A) NAME/KEY: sig-peptide (B) LOCATION: 1-57 (ix) FEATURE:
(A) NAME/KEY: mat_peptide (B) LOCATION: 58-429 (ix) FEATURE:
(A) NAME/KEY: CDS
(B) LOCATION: 1-429 (ix) FEATURE:
(A) NAME/KEY: misc_feature (B) LOCATION: 1 (D) OTHER INFORMATION: /note= "pBAG195 insert: AS heavy chain variable region"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:5:

Met Asp Trp Thr Trp Arg Val Phe Cys Leu Leu Ala Val Ala Pro Gly Ala His Ser Gln Val Gln Leu Gin Glu Ser Gly Pro Gly Leu Val Arg Pro Ser Gln Thr Leu Ser Leu Thr Cys Thr Ala Ser Gly Phe Asn Ile Lys Asp Thr Tyr Met His Trp Val Arg Gln Pro Pro Gly Arg Gly Leu Glu Trp Ile Gly Arg Ile Asp Pro Ala Ser Gly AMENDED SHEET

_2155303 PCT/tiS94!01456 - 56 - 09/02).'1994 45 5~ -`.._ .=C - vAC CCG `__G --. CIG ~-1. ^C~ ..
- .. .. . _,. -C- F"'G .. _ G 7 -s= = s =sp ?,-õ Lys ?^e ~_n Va= Thr Met Le .->;;C .iGC Aal.C C"v AGC .. T .. Al: '': _ ~ C,." ' ,GC :,-Gl. " vTG - 15 ..
a_ r_sti :'~i'_' S e r Ser As:: GL-. ?^e Ser L2L: r._ Val .. L S2r Ser v7- 80 55 l_ Jl... G1.... L.'.= `:1. L:T` _~_ _~_ _=.- L.. LiJ vG - v ^-G 3=
^.h_ ,-_n .--a =.sv T'_ *-_a Va_ _ _ '^-,=_ ^;/s - _._d :aa G_y Mez _ 9C .= _OC
_'_'A TC:- .yCG GG,'m TAT GC'2' C"_'G GP-.C '?^C TGG GGC C_~= GGG ACC =.CG
e_05 d~ Ser T ti,-_ G-`/ ?y~ a Leu Asp Phe T^ G=v G1r G_-/ ':'hr _' r .:'"C ACC GTC '^C.. TCl GGT GAG TC.
r 7 ~
'va''.~ ^'hr Val Se: Ser G 1V G~ u Ser (2) _NFOR.MATION FOR SEQ ?D NO:6:
S E QUgNCE C.ARF.CTE'RIST2CS :

(Ai LENGTH: 143 amino acids (3) TYPE: amino acid (D) TOPOLOGY: linear i_ ) MOLECLZE 'T'YPE : protein (xi) SEQUENCE DESCRi?TiON: SEQ ;D NC:5:

Met -ksp ':'rp '"hr Trp Arc Va-'.~ ?he Cvs Leu Leu Ala Val Ala ?-o -15 -L0 _5 Glv r_la _is Ser Gin Va_ Gin Leu GLn GSer Glv ?rõ G_v Leu -=ia'_ ..-~ P-o Ser G! n Thr Le'.. Se= Leu ^'hr Cvs Thr --~a Ser G_V

-~ 2C 25 e ys^ _-e LVs -s? Thr Tvr Met ..-s __p Val . GL. _ D.-o J~
_C _ 30 Z~ -C

21v a-r SIv Le:. "_':p T'~_e G='/ ~_rc _'_e a.sp ?r. A-_a Ser G_ v ?.SD Th= L; s?'V= Asv ?r.. 7VS ? le G_' Vc1 _ Jd_ :'~:r '~Ie~ Le i Va'_ Asp ''r Ser Ser Asn Gin Phe Ser L,eu Arc Leu Ser Ser Val AMENDED Sr1LET

_2155303 Thr Ala Ala Asp Thr Ala Val Tyr Tyr Cys Ala Asp Gly Met Trp Val Ser Thr Gly Tyr Ala Leu Asp Phe Trp Gly Gln Gly Thr Thr Val Thr Val Ser Ser Gly Glu Ser (2) INFORMATION FOR SEQ ID NO:7:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 386 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA

(ix) FEATURE:
(A) NAME/KEY: sig_peptide (B) LOCATION: 1-57 (ix) FEATURE:
(A) NAME/KEY: mat_peptide (B) LOCATION: 58-386 (ix) FEATURE:
(A) NAME/KEY: CDS
(B) LOCATION: 1-386 (ix) FEATURE:
(A) NAME/KEY: misc-feature (B) LOCATION: 1 (D) OTHER INFORMATION: /noteffi "pBAG198 insert: VK2 (SVMDY) light chain variable region"

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:7:

Met Gly Trp Ser Cys Ile Ile Leu Phe Leu Val Ala Thr Ala Thr Gly Val His Ser Ser Ile Val Met Thr Gln Ser Pro Ser Ser Leu Ser Ala Ser Val Gly Asp Arg Val Thr Ile Thr Cys Lys Ala Ser AMENDED SF',EET

_2155303 Gln Ser Val Thr Asn Asp Val Ala Trp Tyr Gln Gln Lys Pro Gly Lys Ala Pro Lys Leu Leu Ile Tyr Tyr Ala Ser Asn Arg Tyr Thr Gly Val Pro Asp Arg Phe Ser Gly Ser Gly Tyr Gly Thr Asp Phe Thr Phe Thr Ile Ser Ser Leu Gln Pro Glu Asp Ile Ala Thr Tyr Tyr Cys Gln Gln Asp Tyr Ser Ser Pro Tyr Thr Phe Gly Gln Gly Thr Lys Val Glu Ile Lys Arg Lys (2) INFORMATION FOR SEQ ID NO:8:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 128 amino acids (B) TYPE: amino acid (D) TOPOLOGY: linear (ii) MOLECULE TYPE: protein (xi) SEQUENCE DESCRIPTION: SEQ ID NO:8:

Met Gly Trp Ser Cys Ile Ile Leu Phe Leu Val Ala Thr Ala Thr Gly Val His Ser Ser Ile Val Met Thr G1n Ser Pro Ser Ser Leu Ser Ala Ser Val Gly Asp Arg Val Thr Ile Thr Cys Lys Ala Ser Gln Ser Val Thr Asn Asp Val Ala Trp Tyr Gln Gln Lys Pro Gly Lys Ala Pro Lys Leu Leu Ile Tyr Tyr Ala Ser Asn Arg Tyr Thr AiviENDELD SHEET

_2155303 Gly Val Pro Asp Arg Phe Ser Gly Ser Gly Tyr Gly Thr Asp Phe Thr Phe Thr Ile Ser Ser Leu Gln Pro Clu Asp Ile Ala Thr Tyr Tyr Cys Gln Gln Asp Tyr Ser Ser Pro Tyr Thr Phe Gly Gln Gly Thr Lys Val Glu Ile Lys Arg Lys (2) INFORMATION FOR SEQ ID N0:9:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 1348 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA

(ix) FEATURE:
(A) NAME/KEY: VCAM-1 gene segment (B) LOCATION: 1-219 (D) OTHER INFORMATION: This portion of the sequence corresponds, in part, to Exons I, II and III
nucleotide sequence of the VCAM-1 gene of Cybulsky et al. Proc. Nat'l. Acad. Sci. USA 88:7861 (1991).
(ix) FEATURE:
(A) NAME/KEY: Hinge region (B) LOCATION: 220-229 (D) OTHER INFORMATION: This portion of the sequence corresponds, in part, to Fig. 12A in PCT/US92/
0250 and represents the hinge region of Human IgGl heavy chain constant region.

(ix) FEATURE:
(A) NAME/KEY: Heavy chain constant region 2 (B) LOCATION: 230-338 (D) OTHER INFOR.NIATION: This portion of the sequence corresponds, in part, to Fig. 12A in PCT/US92/
02050 and represents the heavy chain constant region 2 of Human IgGl heavy chain constant region.

(ix) FEATURE:
(A) NAME/KEY: heavy chain constant region 3 (B) LOCATION: 339-446 ~ Iu !'t~JCv SHCC

_2155303 (D) OTHER INFORMATION: This portion of the sequence corresponds, in part, to Fig. 12A in PCT/US92/
02050 and represents the heavy chain constant region 3 of Human IgGl heavy chain constant region.

Met Pro Gly Lys Met Val Val Ile Leu Gly Ala Ser Asn Ile Leu Trp Ile Met Phe Ala Ala Ser Gln Ala Phe Lys Ile Glu Thr Thr Pro Glu Ser Arg Tyr Leu Ala Gln Ile Gly Asp Ser Val Ser Leu Thr Cys Ser Thr Thr Gly Cys Glu Ser Pro Phe Phe Ser Trp Arg Thr Gln Ile Asp Ser Pro Leu Asn Gly Lys Val Thr Asn Glu Glv Thr Thr Ser Thr Leu Thr Met Asn Pro Val Ser Phe Gly Asn Glu His Ser Tyr Leu Cys Thr Ala Thr Cys Glu Ser Arg Lys Leu Glu Lys Gly Ile Gln Val Glu Ile Tyr Ser Phe Pro Lys Asp Pro Glu Ile His Leu Ser Gly Pro Leu Glu Ala Gly Lys Pro Ile Thr Val Lys Cys Ser Val Ala Asp Val Tyr Pro Phe Asp Arg Leu Glu Ile Asp Leu Leu Lys Gly Asp His Leu Met Lys Ser Gln Glu Phe Leu AMEhDED SHEET

_2155303 Glu Asp Ala Asp Arg Lys Ser Leu Glu Thr Lys Ser Leu Glu Val Thr Phe Thr Pro Val Ile Glu Asp I1e Gly Lys Val Leu Val Cys Arg Ala Lys Leu His Ile Asp Glu Met Asp Ser Val Pro Thr Val Arg Gln Ala Val Lys Glu Leu Gln Val Asp Lys Thr His Thr Cys Pro Pro Cys Pro Ala Pro Glu Leu Leu Gly Gly Pro Ser Val Phe Leu Phe Pro Pro Lys Pro Lys Asp Thr Leu Met Ile Ser Arg Thr Pro Glu Val Thr Cys Val Val Val Asp Val Ser His Glu Asp Pro Glu Val Lys Phe Asn Trp Tyr Val Asp Gly Val Glu Val His Asn Ala Lys Thr Lys Pro Arg Glu Glu Gln Tyr Asn Ser Thr Tyr Arg Val Val Ser Val Leu Thr Val Leu His Gln Asp Trp Leu Asn Gly Lys Glu Tyr Lys Cys Lys Val Ser Asn Lys Ala Leu Pro Ala Pro Ile Glu Lys Thr Ile Ser Lys Ala Lys Gly Gln Pro Arg Clu Pro ~r Gln Val Tyr Thr Leu Pro Pro Ser Arg Asp Glu Leu Thr Lys Asn G1n Val Ser Leu Thr Cys Leu Val Lys Gly Phe Tyr Pro Ser Asp Ile Ala Val Glu Trp Glu Ser Asn Gly Gln Pro Glu Asn Asn Tyr Lys Thr Thr Pro Pro Val Leu Asp Ser Asp Gly Ser Phe Phe,Leu Tyr Ser Lys Leu Thr Val Asp Lys Ser Arg Trp Gln Gln Gly Asn Val Phe Ser Cys Ser Val Met His Glu Ala Leu His Asn His Tyr Thr Gln Lys Ser Leu Ser Leu Ser Pro Gly Lys (2) INFORMATION FOR SEQ ID N0:10:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 24 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA

(ix) FEATURE:
(A) NAME/KEY:
(B) LOCATION:
(D) OTHER INFORMATION: This corresponds to Kinase Primer 370-31.

(xi) SEQUENCE DESCRIPTION: SEQ ID N0:10:

Asp Lys Thr His Thr Cys ~~'=~T
AMENDED

(2) INFORMATION FOR SEQ ID N0:11:
(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 24 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA

(ix) FEATURE:
(A) NAME/KEY:
(B) LOCATION:
(D) OTHER INFORMATION: This corresponds to Kinase Primer 370-32.

(xi) SEQUENCE DESCRIPTION: SEQ ID N0:11:

(2) INFORMATION FOR SEQ ID NO:12:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 115 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA

(ix) FEATURE:
(A) NAME/KEY:
(B) LOCATION:
(D) OTHER INFORMATION:

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:12:

(2) INFORMATION FOR SEQ ID N0:13:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 39 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear Ai':r,`N I-] E J ^t =`

(ii) MOLECULE TYPE: cDNA
(ix) FEATURE:
(A) NAME/KEY:
(B) LOCATION:
(D) OTHER INFORMATION:

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:13:

Met Pro Gly Lys Met Val Val (2) INFORMATION FOR SEQ ID NO:14:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 23 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA

(ix) FEATURE:
(A) NAME/KEY:
(B) LOCATION:
(D) OTHER INFORMATION:

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:14:

(2) INFORMATION FOR SEQ ID NO:15:

(i) SEQUENCE CHARACTERISTICS:
(A) LENGTH: 14 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA

n'llti.,_.~~J~=~ Ji~._~ i _2155303 (ix) FEATURE:
(A) NAME/KEY:
(B) LOCATION:
(D) OTHER INFORMATION:

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:15:

.-_. = , r; ~~

Claims (26)

CLAIMS:

1. Use of an antibody against the .alpha.4 subunit of very late antigen 4 (VLA-4), or an antigen-binding fragment thereof, for the manufacture of a medicament for prevention of insulin dependent type I diabetes in a prediabetic individual.

2. Use according to claim 1, wherein the antibody is selected from the group consisting of HP1/2, HP2/1, HP2/4, L25, and P4C2.

3. Use according to claim 1, wherein the antibody is HP1/2.

4. Use according to any one of claims 1 to 3, wherein the antibody is a humanized or chimeric antibody.

5. Use according to any one of claims 1 to 4, wherein the medicament is in a dosage form adapted to provide from about 0.1 to about 10 mg/kg of the antibody, based on the weight of the prediabetic individual.

6. Use according to any one of claims 1 to 5, wherein the medicament is in an amount effective to coat VLA4-positive cells in the peripheral blood for a period of 1-14 days.

7. Use according to any one of claims 1 to 6, wherein the medicament is in an amount effective to provide a plasma level of antibody in the prediabetic individual of at least 1 µg/ml.

8. Use according to any one of claims 1 to 7, wherein the medicament is adapted for administration prior to the development of overt diabetes, as measured by a serum glucose level of less than about 250 mg/dL.

9. Use according to any one of claims 1 to 8, wherein the prediabetic individual is a human.

10. Use of an antibody, a recombinant antibody, a chimeric antibody, or antigen-binding fragment of such antibodies that bind to the .alpha.4 subunit of very late antigen 4 (VLA4) for the manufacture of a medicament for treatment of insulin dependent type I diabetes in a mammal.

11. Use according to claim 10, wherein the antigen-binding fragment is selected from Fab, Fab', F(ab')2 and F(v) fragments.

12. Use according to claim 10 or 11, wherein the medicament comprises a plurality of the antibodies or antigen-binding fragments thereof.

13. Use according to any one of claims 10 to 12, wherein the medicament is in a dosage form adapted to provide from about 0.1 to about 10 mg/kg of antibody or antibody fragment, based on the weight of the mammal.

14. Use according to any one of claims 10 to 13, wherein the medicament is in an amount effective to coat VLA4-positive cells in the peripheral blood for a period of 1-14 days.

15. Use according to any one of claims 10 to 14, wherein the medicament is in an amount effective to provide a plasma level of antibody in the mammal of at least 1 µg/ml over a period of 1-14 days.

16. Use of a soluble vascular cell adhesion molecule-1 (VCAM-1) polypeptide for the manufacture of a medicament for treatment of insulin dependent type I diabetes.

17. Use according to claim 16, wherein the soluble VCAM-1 polypeptide is a VCAM-1/Ig fusion protein.

18. A pharmaceutical composition for use in inhibiting onset of insulin dependent type I diabetes, wherein the composition consists essentially of a monoclonal antibody recognizing the .alpha.4 subunit of very late antigen 4 (VLA4) in a pharmaceutically acceptable carrier.

19. The pharmaceutical composition according to claim 18, wherein the monoclonal antibody is a humanized antibody.

20. Use of an antibody against the .alpha.4 subunit of very late antigen 4 (VLA-4), or an antigen-binding fragment thereof, for the treatment of insulin dependent type I
diabetes.

21. Use according to claim 20, wherein the antibody is selected from the group consisting of HP1/2, HP2/1, HP2/4, L25, and P4C2.

22. Use according to claim 20 or 21, wherein the antibody is a humanized or chimeric antibody.

23. Use according to claim 20, wherein the antigen-binding fragment is selected from Fab, Fab', F(ab')2 and F(v) fragments.

24. Use of a soluble vascular cell adhesion molecule-1 (VCAM-1) polypeptide for the treatment of insulin dependent type I diabetes.

25. Use according to claim 24, wherein the soluble VCAM-1 polypeptide is a VCAM-1/Ig fusion protein.

26. Use according to claim 17 or 25, wherein the soluble VCAM-1/Ig fusion protein is VCAM 2D-IgG.